>Perhaps well-funded scientific groups should start buying all these hobbyist GPU farms while they're being sold cheap and put them to use folding proteins or m

>Perhaps well-funded scientific groups should start buying all these hobbyist GPU farms while they're being sold cheap and put them to use folding proteins or modelling whatever? This seems likely. I don't imagine that FPGA clusters depreciate much over periods of a few years. I bet you could make most of the investment back by selling them off to some research firm once the return from the mining dropped. There must be plenty of hi-tech research based startups that need to do serious number crunching on a budget.